CN1504793A - Three dimensional stochastic screen for LCD and video - Google Patents
Three dimensional stochastic screen for LCD and video Download PDFInfo
- Publication number
- CN1504793A CN1504793A CNA200310120761A CN200310120761A CN1504793A CN 1504793 A CN1504793 A CN 1504793A CN A200310120761 A CNA200310120761 A CN A200310120761A CN 200310120761 A CN200310120761 A CN 200310120761A CN 1504793 A CN1504793 A CN 1504793A
- Authority
- CN
- China
- Prior art keywords
- pixel
- value
- blank
- trooping
- array
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
- H04N1/40—Picture signal circuits
- H04N1/405—Halftoning, i.e. converting the picture signal of a continuous-tone original into a corresponding signal showing only two levels
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Control Of Indicators Other Than Cathode Ray Tubes (AREA)
- Image Processing (AREA)
- Controls And Circuits For Display Device (AREA)
Abstract
A dither matrix generator defines a three-dimensional array of pixels referred to as the 'cluster array' in which each pixel is designated based on its value as a 'cluster pixel' or a 'void pixel'. The dither matrix generator also defines a three-dimensional array referred to as the 'dither matrix,' and assigns rank values to the dither matrix in two stages. In the first stage, the dither matrix generator identifies the largest cluster in the cluster array, identifies a cluster pixel located in the largest cluster, assigns a rank value to the element in the dither matrix corresponding to the cluster pixel and removes the cluster pixel from the cluster array. The process is repeated until the cluster array contains no cluster pixels. In the second stage, the dither matrix generator resets the pixels in the cluster array to their initial values, identifies the largest void in the cluster array, identifies a void pixel within the largest void, assigns a rank value to the element in the dither matrix corresponding to the void pixel and inserts a cluster pixel in the location of the void pixel. The process is repeated until the cluster array contains only cluster pixels.
Description
Technical field
The present invention relates to the processing of view data, be used to strengthen the outward appearance of the 3 d image data that provides with visual form or provide by the device such as LCD (" LCD ").
Background technology
Many being designed to can't make full use of the color information that provides the view data from the device of grating images data generation coloured image.Some devices limitation that possesses skills makes them can't generate the color of the four corner of representing in the view data.For example, in some grating images data, the color information of each pixel comprises and is used for for example data of these three kinds of component colors of red, green, blue that the brightness of each color allows 256 kinds of possible brightness degrees by eight bit representations for every kind of component color in these three kinds of colors.If Biao Shi image is presented at and can only produces on the device of eight different brightness degrees each component color by this way, then lose a large amount of color informations.In these cases, adopt quantization method usually, color information quantized, allow this device use that the color that provides in the view data is provided and come display image as uniform quantization.Except eliminating from image the color information, quantization method usually causes occurring in the output image visual artefact.For example, if the quantification of view data causes the visible transformation between the color levels, the effect of crispening then may appear being called.
The needs of optimizing the color function of display device not only appear in the demonstration of two dimension (" 2-D ") view data, and appear in three-dimensional (" 3-D ") view data, the demonstration as video.Term used herein " 3 d image data " is meant Digital Image Data, and wherein the color of pixel is not only the function of two volume coordinates, but also is the function of time coordinate.When 3 d image data be presented on display device, as LCD or computer monitor on the time, the single pixel on the screen may be passed in time and be shown different color.Pel array with identical time coordinate generally is called " frame ".Usually show with rapid succession or " circulation " multiframe, thereby produce live image.
A kind of technology that is used to expand the limited obtainable Color Range of device of color function is called three-dimensional linked network.In general, screening technology utilizes the characteristic of naked eyes to come the color output that exceeds its intrinsic color function of expanding unit.If the figure that naked eyes watch enough little point with different colours to constitute, the beholder generally can not feel the color of each point, but feels the color that approximates the average color of each point in the figure.When eyes were observed the little array of the point on the display device, the ability of human eye perception average color spatially worked, and when eyes are observed the pel array of the rapid change of its color, then passed in time and worked.In either case, eyes are felt a kind of color of the average color that approximates observed pixel usually.The color that the beholder experienced is called " effectively color output ".Be used to produce cubical array or " screen (screen) " of the common structured value of screening technology of live image, they are used in and produce the color figure with required effective color output in the pel array.These technology seem to have the middle extra color levels of the defined color levels of physical characteristics of device by generation, make device can expand its displayable number of colours.
The common three-dimensional screening technology that is called frame frequency modulation (" FRM ") alternately switches on and off the pixel on the multiframe, produces the effective color output that approximates the average color of measuring on a succession of frame.Term " refresh rate " is meant the frequency of output conversion between the different color grade of single pixel.Improve refresh rate and can make device can show the Neutral colour of greater number, and the bigger control that effective color is exported can be provided.
Another factor that influences the perceived quality of color output is the perception homogeneity of color graphics.When presenting 3 d image data, preferably use the screen that produces color graphics with high homogeneity.It is inconsistent that known FRM technology produces on the ability of even figure at it.
Need a kind of improved three-dimensional screening technology of exploitation, it increases device producible color levels quantity when presenting 3 d image data, and the inhomogeneity bigger control to shown color graphics is provided, and artifactitious the appearance minimized.
Summary of the invention
An object of the present invention is by producing three-dimensional screen, display device can be produced be used to present a plurality of color levels of 3 d image data from any three-dimensional colour figure.
The definition of initial graphics generator is called the cubical array of the pixel of " initial graphics ".Adopt initial graphics as starting point, evenly the pattern generator definition is called the cubical array of the pixel of " cluster array ", the maximum that adopts wave filter to discern in the cluster array is trooped, elimination is positioned at the pixel of trooping that maximum cluster is concentrated, maximum blank in the identification cluster array, and the pixel of will trooping is placed in the maximum blank.Evenly pattern generator constantly will troop pixel from maximum troop move to maximum blank, maximum blank up to concentrate the step of eliminating the pixel of trooping to create from maximum cluster.When this situation occurring, finish this process, and the cluster array of revising is called even figure.The definition of color lump (dither) matrix generator is called the cubical array of " color lump matrix ", and two stages rank value is distributed to the color lump matrix.In the phase one, the maximum that dither matrix generator is discerned in the even figure is trooped, and identification is positioned at the pixel of trooping that maximum cluster is concentrated, and rank value is distributed in the color lump matrix element corresponding to the pixel of trooping.And then from even figure, eliminate the pixel of trooping.Repeat this process, do not comprise any pixel of trooping up to even figure.In subordinate phase, dither matrix generator resets to its initial value with the pixel in the even figure, discerns the maximum blank in the even figure, discerns the blank pixel in the maximum blank, and the position of the pixel insertion blank pixel of will trooping.Dither matrix generator is distributed in the color lump matrix element corresponding to blank pixel with rank value.Repeat this process, only comprise the pixel of trooping up to even figure.
Description of drawings
Fig. 1 illustrates the critical piece in the typical image reproducing system.
Fig. 2 illustrates the critical piece among the typical personal computer system.
Fig. 3 is the block diagram that can be used for carrying out a kind of system of each side of the present invention.
Fig. 4 (a)-4 (c) illustrates three frames of the random figure that can be used as initial graphics.
Fig. 4 (d) illustrates by the pixel value in the frame shown in 4 (a)-4 (c) is averaged the image that produces.
Fig. 5 illustrates a process flow diagram, sums up by even pattern generator and the performed step of dither matrix generator.
The two-dimentional cluster array of Fig. 6 (a) illustrative binary value.
Cluster array shown in a kind of Fig. 6 of can be applicable to of Fig. 6 (b) illustrative (a) is trooped or maximum blank wave filter to discern maximum.
Fig. 6 (c) illustrative is duplicated the cluster array of Fig. 6 (a) of four times.
Fig. 6 (d) illustrates filter applies shown in Fig. 6 (b) in the details in the zone of the cluster array of Fig. 6 (a) with numeral.
The array of the total density value of pixel in the cluster array shown in Fig. 6 (e) illustrative Fig. 6 (a).
Fig. 7 (a)-7 (c) illustrates from three frames of the even figure of the generation of initial graphics shown in Fig. 4 (a)-4 (c).
Fig. 7 (d) illustrates by the value to pixel in the frame shown in Fig. 7 (a)-7 (c) and averages the image that produces.
Fig. 8 (a) and 8 (b) illustrate the example value of Gaussian filter.
Fig. 9 (a) and 9 (b) illustrate a kind of example value of wave filter.
Figure 10 (a)-10 (c) illustrates from three frames of the even figure of the generation of initial graphics shown in Fig. 4 (a)-4 (c).
Figure 10 (d) illustrates by the value to pixel in the frame shown in Figure 10 (a)-10 (c) and averages the image that produces.
Figure 11 (a)-11 (c) illustrates from three frames of the even figure of the generation of initial graphics shown in Fig. 4 (a)-4 (c).
Figure 11 (d) illustrates by the value to pixel in the frame shown in Figure 11 (a)-11 (c) and averages the image that produces.
Figure 12 (a)-12 (c) illustrates from three frames of the even figure of the generation of initial graphics shown in Fig. 4 (a)-4 (c).
Figure 12 (d) illustrates by the value to pixel in the frame shown in Figure 12 (a)-12 (c) and averages the image that produces.
Embodiment
System survey
Fig. 1 illustrates the critical piece in the typical image reproducing system.Input media 10 receives the signal of representing original image from path 1, and represents along the rasterisation that path 11 generates original images.Control device 20 receives these expressions and responds from path 11, generates the relevant expression of output unit of original images along path 21.Output unit 30 receives these expressions and responds from path 21, generates the expression of original images along path 31.The present invention is intended to improve the perceived quality of the expression that output unit produces.
Fig. 2 explanation can be used for carrying out the critical piece in a kind of typical personal computer system of each side of the present invention.CPU 42 provides computational resource.The interface of I/O control 43 expressions and the I/O device 44 such as keyboard, mouse or modulator-demodular unit.RAM45 is a system random access memory.The interface of storage control 46 expressions and memory storage 47, memory storage 47 comprises medium or the optical media such as tape or disk.Medium can be used to the program of recording operation system, utility routine and the used instruction of application program, and can comprise the program that realizes each side of the present invention.Demonstration control 48 provides the interface with display device 49.Display device can be the visual display unit of any kind.The interface of control 50 expressions and the input media 10 such as software application that can produce graph image or video camera.The interface of control 52 expressions and the output unit 30 such as the LCD device.
Among Fig. 2, all main system units all are connected with the bus 41 that can represent more than a kind of physical bus.Do not need bus architecture to implement the present invention.
The function of one or more machine elements and various aspects of the present invention can realize by various circuit, comprising discreet logic parts, one or more ASIC and/or software-controlled processor.For example, control device 20 can be realized by isolated plant.The mode that realizes control device 20 is not very important for the present invention.Can adopt other realization that comprises numeral and analog processing circuit.Can or comprise any basically magnetic of employing or the optical recording technology is transmitted information, the medium that comprises tape, disk and CD waits and transmits software of the present invention and realize by the various machine-readable mediums base band on the frequency spectrum that comprises from superaudio to ultraviolet frequencies or the modulation communication path.
Fig. 3 is the block diagram that can be used for carrying out a kind of system of each side of the present invention.Initial graphics generator 510 produces the data of the figure of expression value, and sends data to even pattern generator 520 along path 515.Evenly pattern generator 520 is handled the data that receive along path 515, and sends data to dither matrix generator 540 along path 525.Dither matrix generator 540 adopts the data that receive along path 525 to generate three-dimensional screen, and will represent that along path 545 data of screen send to memory storage 47.Halftone process device 729 receives the 3 d image data of rasterisations along path 740, the data of expression screen in the access to storage device 47, and send the halftone representation of image to output unit 30 along path 744.Output unit 30 produces output image along path 753.In best a realization, initial graphics generator 510, even pattern generator 520 and dither matrix generator 540 realize by one or more software application that are combined in the control device 20.Another kind of scheme is that initial graphics generator 510, even pattern generator 520 and dither matrix generator 540 realize by the numeral or the mimic channel that are combined in the control device 20.In another is realized, initial graphics generator 510, even pattern generator 520 and dither matrix generator 540 are combined in the output unit 30, and can or can be realized by the software application that numeral or mimic channel are carried out by output unit 30 by one or more.In another was realized, initial graphics generator 510, even pattern generator 520 and dither matrix generator 540 were combined in the control device 20, and memory storage 47 and halftone process device 729 are combined in the output unit 30.
The initial graphics generator
Even pattern generator
Fig. 5 illustrates a process flow diagram, sums up by the performed step of even pattern generator 520 and dither matrix generator 540.Step 550-557 is carried out by even pattern generator 520.In the best realized, evenly pattern generator 520 definition M * N * T arrays were called " cluster array ".In step 550, evenly pattern generator 520 copies to cluster array by the value of respective pixel in each pixel distribution initial graphics in cluster array with initial graphics.
The identification that maximum is trooped
Evenly pattern generator 520 is converted to cluster array equally distributed figure in such a way.In step 551, evenly the maximum in the pattern generator 520 identification cluster arrays is trooped.In the best realized, maximum was trooped and is discerned by wave filter.If adopt X * Y * Z wave filter, then drag X * Y * Z filtering window by three-dimensional cluster array, so that each X * Y of the neighbor in the check cluster array * Z set or zone individual element.Evenly pattern generator 520 is filter applies to a zone, carries out convolution by pixel value and filter function in should the zone, thus the generation output valve.Select the pixel in this zone and be appointed as " center pixel " in this zone.The selection of zone center pixel can be arbitrarily, perhaps also can be determined by filter characteristic or other factors.The output valve that produces from respective regions is called " total density value ", and relevant with the center pixel in zone.Two dimension 6 * 6 cluster arrays 705 of Fig. 6 (a) illustrative binary value.Comprise pixel 723 and have " 0 " value in interior one part of pixel.Comprise pixel 724 and have " 1 " value in other interior pixel.Cluster array 705 shown in a kind of Fig. 6 of can be applicable to of Fig. 6 (b) illustrative (a) is to discern 3 * 3 wave filters that maximum is trooped.Although in Fig. 6 (a) and 6 (b), provide two-dimentional cluster array 705 and two dimensional filter 710 for convenience of explanation and respectively, the process that three-dimensional filter is applied to three-dimensional cluster array with two-dimensional array in identical mode carry out.In case of necessity, cluster array is repeatedly duplicated, so that allow wave filter to produce the total density value of boundary pixel by unrolling.Fig. 6 (c) illustrative is duplicated four times cluster array 705, is the situation in zone 727 of nine pixels at center so that explanation is applied to wave filter 710 with pixel 723.Wave filter 710 carries out convolution with the value in the zone 727, produces the total density value of center pixel 723.Fig. 6 (d) is applied to wave filter 710 with the numeral explanation situation in zone 727.With reference to the digital computation shown in Fig. 6 (d), zone 727 has total density value 2.2.Wave filter 710 is applied to each 3 * 3 zone in the cluster array 705.The array 733 of the total density value of Fig. 6 (e) illustrative cluster array 705.Each value in the array 733 is corresponding to the respective pixel in the cluster array 705, and expression is the total density value in the zone at center with the respective pixel.For example, element 736 is corresponding to pixel 723, and comprises value 2.2, its expression and regional 727 relevant total density values.Element 738 is corresponding to pixel 724, and comprises value 4.5, and its expression is the total density value in the zone at center with pixel 724.
Identification has the pixel in the cluster array of maximum total density value, and then circle zone is confirmed as maximum and troops.With reference to Fig. 6 (e), the 4.5th, the maximum total density value in the array 733 shows with respective pixel 724 to be that the zone at center is that maximum in the cluster array 705 is trooped.
In step 552, evenly pattern generator 520 is eliminated and is positioned at the pixel of trooping that maximum cluster is concentrated the centre.In the best realized, the elimination of the pixel of trooping was to be positioned at the pixel and change its value into 0 from 1 and realize of trooping that maximum cluster is concentrated the centre by identification.If the center pixel that maximum cluster is concentrated is a blank pixel, then has second largest the trooping of the pixel of trooping the elimination pixel of trooping from the center.In another was realized, can eliminate maximum cluster concentrated was not the pixel that is positioned at central authorities.
Maximum blank identification
In step 553, evenly the maximum in the pattern generator 520 identification cluster arrays is blank.In the best realizes, be used to discern the wave filter that maximum troops and also be used for discerning maximum blank.Check each regional total density value in the cluster array, the zone with minimum total density value is identified as maximum blank.
In step 555, the pixel of trooping is placed in the maximum blank.In the best realizes, be positioned at the blank pixel of the white central authorities of maximum empty and change its value into 1 from 0 by identification, the pixel of will trooping is placed in the maximum blank.If maximum blank center pixel is the pixel of trooping, the pixel center of being placed on of then will trooping has in the second largest blank of blank pixel.In another was realized, can eliminate in the maximum blank was not the pixel that is positioned at central authorities.
If it is just from wherein eliminating the zone of the pixel of trooping that step 554 is determined maximum blank, then step 556 pixel of will trooping returns to its position in zone central authorities, finishes the process that constitutes cluster array.Otherwise evenly pattern generator 520 turns back to step 551, and checks amended cluster array, and the identification maximum is trooped, and repeats this process.Evenly pattern generator 520 pixel that continues to troop is trooped from maximum and moved to the maximum blank, and is maximum blank up to concentrate the step of eliminating the pixel of trooping to create from maximum cluster.When this situation occurring, this method is finished this process by entering step 556 as mentioned above.The cluster array of revising is called even figure.
Fig. 7 (a)-7 (c) illustrates from first frame 782, second frame 783 and the 3rd frame 784 of the even figure of the generation of initial graphics shown in Fig. 4 (a) 4 (c).Fig. 7 (d) illustrates by the value in frame 782,783 and 784 is averaged the image that is produced.In step 557, evenly figure is stored in the storer as required.
Dither matrix generator
Step 560-576 is carried out by dither matrix generator 540.In the best realized, dither matrix generator 540 received the data of the even figure of expression from path 525.In this realization, dither matrix generator 540 definition M * N * T cluster arrays, it is used for the cluster array identical functions used with even pattern generator 520.In step 560, dither matrix generator 540 is by the value of respective pixel in each the pixel distributed uniform figure in cluster array, with even graph copying in cluster array.Have been found that and adopt even figure evenly to distribute very favourable because of it as the starting point of dither matrix generator 540.In other was realized, different figures can be loaded in the cluster array in the dither matrix generator 540.In step 561, dither matrix generator 540 is set up variable R ANK, and it shows the current quantity that appears at the pixel of trooping in the cluster array.For example, in the best realizes,, then distribute the initial value that equals Q for RANK if evenly figure comprises Q the pixel of trooping.
Divide two stages that value is distributed to element in the color lump matrix.In the phase one, according to step 563 and 564, the maximum in the dither matrix generator 540 identification cluster arrays is trooped, and identification is arranged in the pixel that this is trooped, and the rank value that will equal variable R ANK is distributed to corresponding to the element in the color lump matrix of this pixel.In step 565, from consideration in the future, eliminate the pixel of trooping.In the best realized, the identification of the pixel of trooping and elimination were to realize by discerning the center pixel that maximum troops and changing its value into 0 from 1.If maximum center pixel of trooping is a blank pixel, then discern and eliminate second largest center pixel of trooping.In step 566, variable R ANK reduces 1, and the pixel of trooping is eliminated in reflection from cluster array.Repeat this process, do not comprise any pixel of trooping up to cluster array, as the judgement in the step 562.This process produces Q unique rank value (dropping to 1 from Q) distributing to the selected element in the color lump matrix.
Subordinate phase will be distributed to element the color lump matrix up to the rank value of M * N * T from Q+1.Subordinate phase is from step 570, wherein dither matrix generator 540 once more with even graph copying in cluster array.In step 571, the value of RANK is reset and is Q, the quantity of the pixel of trooping in the reflection cluster array.In step 573 and 574, the maximum blank in the dither matrix generator 540 identification cluster arrays is discerned the blank pixel in the maximum blank, and insert the pixel of trooping on the position of blank pixel.In the best realizes, by discerning the center pixel in the maximum blank and change its value into 1 from 0, the insertion pixel of trooping.In step 575 and 576, dither matrix generator 540 adds 1 with the value of RANK, and having reflected to cluster array increases the pixel of trooping, and distributes the rank value that equals corresponding to the adjusted value of the RANK of element in the color lump matrix of center pixel.Repeat this process, only comprise the pixel of trooping, as the judgement in the step 572 up to cluster array.The net result of two stage process is the color lump matrix that comprises M * N * T unique value.
Normalization
Said process produces the M * N * T color lump matrix that comprises M * N * T unique rank value.L rank value rather than M * N * T rank value if desired then can be converted to normalization rank value R ' to each rank value R by using following formula:
In following formula, by round up, promptly by casting out the whole numerals that are positioned at radix point the right among the variable x, int{x} produces round values.
The selection of wave filter
The selection of wave filter influences the performance and the outward appearance of 3-D view, therefore is the key element that three-dimensional screen is realized.A kind of Gaussian filter of realizing adopting as giving a definition:
The example value of 5 * 5 pixels more than Fig. 8 (a) and 8 (b) illustrate in two Spatial Dimensions of employing expression and the defined Gaussian filter of 5 * 5 * 3 filtering windows of 3 frames in the time dimension.Fig. 8 (a) is f (x, y, value t) when t=0 is described.Fig. 8 (b) illustrates t=± 1 o'clock f (x, y, value t).With reference to Fig. 7 (a)-7 (c), be applied to the initial graphics shown in Fig. 4 (a)-4 (c) by Gaussian filter with following formula (2) definition, produce frame 782,783 and the 784 even figures that constitute.
Because Gaussian filter is put on an equal footing room and time information, so it can not adapt to the people's who handles room and time information by different way vision system well.Another kind of wave filter at this species diversity is defined by following formula:
Wherein
And α is a real number.
e
-(x2+y2/2 σ 2)Item influences spatially uniform,
Graphics performance in the item influence time dimension.The example value of the wave filter that Fig. 9 (a) and 9 (b) illustrate 5 * 5 * 3 filtering windows that adopt σ=1.0, defined by formula (3) and (4).Fig. 9 (a) is f (x, y, value t) when t=0 is described.Fig. 9 (b) illustrates t=± 1 o'clock f (x, y, value t).In Fig. 9 (b), the value of not specifying α.Fig. 9 (a) produces the different visual patterns that depend on the α value with the wave filter shown in 9 (b).Increase common generation of α value and have the more figure of uniform outer appearance.This result is extremely useful when high refresh rate.Observe, in the various figures of the identical mean flow rate on showing the frame of specified quantity, show that higher inhomogeneity those figures visually feel more comfortable.Figure 10 (a)-10 (c) illustrate from the initial graphics shown in Fig. 4 (a)-4 (c), adopt α=2.0, by three frames 906,907 and 908 of the even figure that wave filter produced of formula (3) and (4) definition.Figure 10 (d) illustrates by cycle frame 906,907 and 908 images that produced.
If α=0.0, filter function do not provide weighting to front or subsequent frame, but respectively each frame is calculated output valve.Three frames 911,912 and 913 of the even figure that Figure 11 (a)-11 (c) illustrates α=0.0 o'clock, produce from initial graphics shown in Fig. 4 (a)-4 (c).Figure 11 (d) illustrates by cycle frame 906,907 and 908 images that produced.
If α＜0, then filter function produces little difference usually or does not have the frame of difference fully.This result is extremely useful when slowly circulating these frames, because the visible difference between the frame may produce undesirable " flicker " when hanging down refresh rate.Three frames 916,917 and 918 of the even figure that Figure 12 (a)-12 (c) illustrates α=-0.6 o'clock, produce from initial graphics shown in Fig. 4 (a)-4 (c).Dot pattern is all identical in whole three frames.Figure 12 (d) illustrates by cycle frame 906,907 and 908 images that produced.
Claims (12)
1. one kind produces three-dimensional color lump matrix to be used for the method for the one or more color levels of presentation video, and wherein said method comprises:
Receive the data of the cubical array of remarked pixel, wherein each pixel has a value, and according to the value of each pixel each pixel is distributed to first set that comprises the pixel of trooping that is arranged in trooping or distributed to second set that comprises the blank pixel that is arranged in the blank;
Definition comprises the three-dimensional color lump matrix of a plurality of elements, and each pixel in the wherein said array has the respective element in the described color lump matrix;
Be arranged in the pixel of trooping that described array maximum cluster concentrates, upgrade the rank value that shows the order of discern the described pixel of trooping, produce the value of respective element described in the described color lump matrix and eliminate the described pixel of trooping, definite value by identification corresponding to each element in the described color lump matrix of the pixel of trooping from described first set according to described rank value; And
By identification be arranged in the maximum blank of described array blank pixel, upgrade the rank value show the order of discern blank pixel, produce the value of respective element described in the described color lump matrix and eliminate described blank pixel, definite value corresponding to each element in the described color lump matrix of blank pixel from described second set according to described rank value.
2. the method for claim 1 is characterized in that, the initial value of described array interior pixel is determined by following operation:
Receive the data of the three-dimensional initial graphics of remarked pixel, wherein each pixel has a value, and according to the value of each pixel each pixel is distributed to first set that comprises the pixel of trooping that is arranged in trooping or distributed to second set that comprises the blank pixel that is arranged in the blank; And
By identification repeatedly be arranged in the pixel of trooping that described initial graphics maximum cluster concentrates, the pixel of will describedly trooping from described first set be re-assigned to described second set, identification is positioned at the blank pixel of the maximum blank of described initial graphics and described blank pixel is re-assigned to described first from described second set and gathers, and adjusts the value of pixel in the described initial graphics.
3. the method for claim 1 is characterized in that, each pixel in the described array has binary value.
4. the method for claim 1, it is characterized in that the step of upgrading described rank value subtracts one and for each blank pixel of eliminating described rank value is added one realize from described second set by the pixel of trooping for each that eliminate with described rank value from described first set.
5. a device comprises storer and the treatment circuit that is connected described storer, and wherein said treatment circuit is suitable for:
Receive the data of the cubical array of remarked pixel, wherein each pixel has a value, and according to the value of each pixel each pixel is distributed to first set that comprises the pixel of trooping that is arranged in trooping or distributed to second set that comprises the blank pixel that is arranged in the blank;
Definition comprises the three-dimensional color lump matrix of a plurality of elements, and each pixel in the wherein said array has the respective element in the described color lump matrix;
Be arranged in the pixel of trooping that described array maximum cluster concentrates, upgrade the rank value that shows the order of discern the described pixel of trooping, produce the value of respective element described in the described color lump matrix and eliminate the described pixel of trooping, definite value by identification corresponding to each element in the described color lump matrix of the pixel of trooping from described first set according to described rank value; And
By identification be arranged in the maximum blank of described array blank pixel, upgrade the rank value show the order of discern blank pixel, produce the value of respective element described in the described color lump matrix and eliminate described blank pixel, definite value corresponding to each element in the described color lump matrix of blank pixel from described second set according to described rank value.
6. device as claimed in claim 5 is characterized in that, the initial value of described array interior pixel is determined by following operation:
Receive the data of the three-dimensional initial graphics of remarked pixel, wherein each pixel has a value, and according to the value of each pixel each pixel is distributed to first set that comprises the pixel of trooping that is arranged in trooping or distributed to second set that comprises the blank pixel that is arranged in the blank; And
By identification repeatedly be arranged in the pixel of trooping that described initial graphics maximum cluster concentrates, the pixel of will describedly trooping from described first set be re-assigned to described second set, identification is positioned at the blank pixel of the maximum blank of described initial graphics and described blank pixel is re-assigned to described first from described second set and gathers, and adjusts the value of pixel in the described initial graphics.
7. device as claimed in claim 5 is characterized in that, each pixel in the described array has binary value.
8. device as claimed in claim 5, it is characterized in that the step of upgrading described rank value subtracts one and for each blank pixel of eliminating described rank value is added one realize from described second set by the pixel of trooping for each that eliminate with described rank value from described first set.
9. one kind is installed readable medium, and it comprises instruction repertorie, and these instructions are carried out by described device, thereby realizes being used for producing three-dimensional color lump matrix to be used for the method for the one or more color levels of presentation video, and wherein said method comprises:
Receive the data of the cubical array of remarked pixel, wherein each pixel has a value, and according to the value of each pixel each pixel is distributed to first set that comprises the pixel of trooping that is arranged in trooping or distributed to second set that comprises the blank pixel that is arranged in the blank;
Definition comprises the three-dimensional color lump matrix of a plurality of elements, and each pixel in the wherein said array has the respective element in the described color lump matrix;
Be arranged in the pixel of trooping that described array maximum cluster concentrates, upgrade the rank value that shows the order of discern the described pixel of trooping, produce the value of respective element described in the described color lump matrix and eliminate the described pixel of trooping, definite value by identification corresponding to each element in the described color lump matrix of the pixel of trooping from described first set according to described rank value; And
By identification be arranged in the maximum blank of described array blank pixel, upgrade the rank value show the order of discern blank pixel, produce the value of respective element described in the described color lump matrix and eliminate described blank pixel, definite value corresponding to each element in the described color lump matrix of blank pixel from described second set according to described rank value.
10. medium as claimed in claim 9 is characterized in that, the initial value of described array interior pixel is determined by following operation:
Receive the data of the three-dimensional initial graphics of remarked pixel, wherein each pixel has a value, and according to the value of each pixel each pixel is distributed to first set that comprises the pixel of trooping that is arranged in trooping or distributed to second set that comprises the blank pixel that is arranged in the blank; And
By identification repeatedly be arranged in the pixel of trooping that described initial graphics maximum cluster concentrates, the pixel of will describedly trooping from described first set be re-assigned to described second set, identification is positioned at the blank pixel of the maximum blank of described initial graphics and described blank pixel is re-assigned to described first from described second set and gathers, and adjusts the value of pixel in the described initial graphics.
11. medium as claimed in claim 9 is characterized in that, each pixel in the described array has binary value.
12. medium as claimed in claim 9, it is characterized in that the step of upgrading described rank value subtracts one and for each blank pixel of eliminating described rank value is added one realize from described second set by the pixel of trooping for each that eliminate with described rank value from described first set.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US10/308,554 US6842184B2 (en) | 2002-12-03 | 2002-12-03 | Three dimensional stochastic screen for LCD and video |
| US10/308554 | 2002-12-03 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1504793A true CN1504793A (en) | 2004-06-16 |
| CN1282003C CN1282003C (en) | 2006-10-25 |
Family
ID=32392777
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2003101207612A Expired - Fee Related CN1282003C (en) | 2002-12-03 | 2003-12-03 | Three dimensional stochastic screen for LCD and video |
Country Status (2)
| Country | Link |
|---|---|
| US (1) | US6842184B2 (en) |
| CN (1) | CN1282003C (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4020132B2 (en) * | 2005-07-21 | 2007-12-12 | セイコーエプソン株式会社 | Dither matrix generation apparatus, dither matrix generation method, program, and image processing apparatus |
| US20070070427A1 (en) * | 2005-08-18 | 2007-03-29 | Lexmark International, Inc. | Systems and methods for selective dithering using pixel classification |
| US7830569B2 (en) * | 2006-03-31 | 2010-11-09 | Eastman Kodak Company | Multilevel halftone screen and sets thereof |
| US7626730B2 (en) | 2006-03-31 | 2009-12-01 | Eastman Kodak Company | Method of making a multilevel halftone screen |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5535020A (en) | 1992-10-15 | 1996-07-09 | Digital Equipment Corporation | Void and cluster apparatus and method for generating dither templates |
| US6147671A (en) | 1994-09-13 | 2000-11-14 | Intel Corporation | Temporally dissolved dithering |
| US6690389B2 (en) * | 1995-04-25 | 2004-02-10 | Compaq Computer Corporation | Tone modifying dithering system |
| US6094187A (en) | 1996-12-16 | 2000-07-25 | Sharp Kabushiki Kaisha | Light modulating devices having grey scale levels using multiple state selection in combination with temporal and/or spatial dithering |
| US6088512A (en) | 1997-07-09 | 2000-07-11 | Seiko Epson Corporation | Void-and-cluster dither-matrix generation for better half-tone uniformity |
| US6008794A (en) | 1998-02-10 | 1999-12-28 | S3 Incorporated | Flat-panel display controller with improved dithering and frame rate control |
| GB2336931A (en) | 1998-04-29 | 1999-11-03 | Sharp Kk | Temporal dither addressing scheme for light modulating devices |
| US6714320B1 (en) * | 1999-06-14 | 2004-03-30 | Toshiba Tec Kabushiki Kaisha | Image processor and color image processor |
| US6791718B1 (en) * | 2000-06-14 | 2004-09-14 | Hewlett-Packard Development Company, L.P. | Halftone printing with dither matrices generated by using cluster filters |
| US6710778B2 (en) * | 2001-02-12 | 2004-03-23 | Lexmark International, Inc. | Method for halftoning using a difference weighting function |
-
2002
- 2002-12-03 US US10/308,554 patent/US6842184B2/en not_active Expired - Fee Related
-
2003
- 2003-12-03 CN CNB2003101207612A patent/CN1282003C/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN1282003C (en) | 2006-10-25 |
| US20040104918A1 (en) | 2004-06-03 |
| US6842184B2 (en) | 2005-01-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8451289B2 (en) | Systems and methods for dither structure creation and application | |
| US6456405B2 (en) | Method and apparatus for displaying computer generated holograms | |
| Wetzstein et al. | Tensor displays: compressive light field synthesis using multilayer displays with directional backlighting | |
| DE69815427T2 (en) | Image display device | |
| DE69822936T2 (en) | Image display device and image evaluation device | |
| EP0463700B1 (en) | Method of and apparatus for generating an image | |
| CN109712165A (en) | A kind of similar foreground picture image set dividing method based on convolutional neural networks | |
| US7352373B2 (en) | Systems and methods for multi-dimensional dither structure creation and application | |
| JP7504120B2 (en) | High-resolution real-time artistic style transfer pipeline | |
| CN113330372B (en) | Holographic display system and method | |
| US8849060B2 (en) | Image processing method, image processing system and display device for modifying background colors based on object colors | |
| CN1282003C (en) | Three dimensional stochastic screen for LCD and video | |
| CN101600039A (en) | The method of half tone image conversion method, Method of printing and generation halftone shield | |
| Maddess et al. | Binary and ternary textures containing higher-order spatial correlations | |
| CN104240195B (en) | Model establishing method and system based on virtual algorithm | |
| CN1725278A (en) | Method and device for processing video data by combining error diffusion and another dithering | |
| DE69506531T2 (en) | FAST FIGURE IN 3-D COMPUTER GRAPHICS | |
| Barbé | Artistic design with fractal matrices | |
| US11303779B2 (en) | Halftone image creation | |
| Heckbert | Techniques for real-time frame buffer animation | |
| JPH0664441B2 (en) | Allocation method of sample points | |
| CN114692776A (en) | Self-adaptive refresh rate field color sequence display driving method, device and related equipment | |
| JPH08336164A (en) | Method for generating image information, method for displaying image and image display system | |
| Lau et al. | Methods of dither array construction employing models of visual perception | |
| Lo et al. | A novel algorithm for color quantization by 3D diffusion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20061025 Termination date: 20131203 |